Chung 1

Koeun Grace Chung Advisor: Richard Larson LIN 495/496 Spring 2011

Usage of Demonstratives 1. Introduction

Deixis is a universal linguistic phenomenon which is at the interface of pragmatics and semantics. It is traditionally divided into three semantic categories: person, place, and time (Diessel 35). Demonstratives, which are the main focus of this paper, are place (or spatial) deictics such as this and that in English. Due to their spatial nature, demonstrative contrasts, i.e. this versus that, are commonly understood with the physical distance of the entities they refer to; the chair at a closer distance might be referred to as this chair while the chair farther away might be referred to as that chair. This, at a first glance, does not seem erroneous, and in fact, it is a correct way of understanding the contrasting usage of this and that. However, it is a simplistic way of viewing the usage of demonstratives.

There are no specific guideline regarding how close or far an object has to be for this or that to be used. No one learns from his kindergarten teacher that an object or person within 3 feet from his or her body is referred with this while everything outside that 3 feet circumference needs to be referred with that. What is more intriguing is that demonstratives are often used for entities or subjects outside of view or even inexistent in the outside world, such as a stolen wallet, a dream, or spoken words. Thus, there must be more than just physical distance that governs the usage of demonstratives, and it is hypothesized that there are other factors such as Common Ground (shared knowledge of the speaker and the addressee) and Source of Idea (whether the idea is originated from the speaker, the addressee, or a third party) in play.

In this study, a specific type of demonstrative will be identified through syntax (section 2), and some of its pragmatic (section 3) and semantic (section 4) features will be described. A survey study (section 5) that examined further questions and hypotheses will be presented: This study considered five different languages – two of which (English and Chinese) with two deictic terms and three of which (Spanish, Korean, and Japanese) with three deictic terms. Finally, findings in the survey will be related back to the pragmatics and semantics of demonstratives (section 6).

Chung 2

2. Syntax

Before delving into the semantics and pragmatics of demonstratives, it is important to define where demonstratives stand in language and, more specifically, which type of demonstratives are to be studied in this paper. We thus begin with a brief mention of demonstrative syntax is an appropriate way to start the discussion.

According to Diessel, demonstratives can be divided into four different grammatical categories: (i) demonstrative pronouns, (ii) demonstrative determiners, (iii) demonstrative adverbs, and (iv) demonstrative identifiers (Diessel 57). Some languages have demonstratives that belong to only one of these categories, while other languages have demonstratives belonging to more than one category. Although demonstrative pronouns and demonstrative determiners are the main interest in this study, all four categories will be mentioned below for the clearer understanding of the two main categories.

2.1 Demonstrative Pronouns

Demonstrative pronouns, also known as “pronominal demonstratives,” function as independent pronouns, like that in (1) and (2).

(1) (referring to a squirrel on the tree) That just threw an acorn at me.

(2) (referring to a book nearby) Can you get that for me?

In (1), that works as a pronoun referring to the subject of the sentence, a squirrel, while that in (2) works as a pronoun referring to the object, a book. Clearly, demonstratives in this category have a function very similar to the nominative pronouns such as I, we, and he and the accusative pronouns such as me, us, and him.

2.2 Demonstrative Determiners

While demonstrative pronouns are independent pronouns, demonstrative determiners, also known as adnominal demonstratives, function as noun modifiers. In most languages, these demonstratives are followed by a co-occurring noun (Diessel 64), as in (3).

(3) This town is very empty.

In many languages, demonstrative determiners appear before noun like an article – compare this in (3) with the article the in (4),

(4) The town is very empty.

Nonetheless, demonstrative determiners and articles belong to two distinct classes with different slots in the noun phrase. This distinction is supported by the fact that demonstrative determiners and articles

Chung 3

co-occur in some languages and that demonstratives and articles follow different word order pattern (Van Valin and LaPolla 62).1

2.3 Demonstrative Adverbs

Most languages do not have distinctive demonstrative forms for demonstrative pronouns and demonstrative determiners, and the status of English demonstratives is rather controversial. However, in some languages such as Japanese, the forms are clearly distinguished by having different stems (Diessel 59), which fortifies the difference between demonstrative pronouns and demonstrative determiners – see more detail on Japanese demonstratives in section 5.2.5.

Unlike the two categories of demonstratives above, demonstrative adverbs are locational deictics such as here and there that function as verb modifiers. It is true that in many languages, locational deictics can be used as a noun modifier, like in English this book here (Diessel 74), in which case the demonstrative adverb functions as a reinforcer to the co-occurring demonstrative determiner. Another subcategory of demonstrative adverbs is manner demonstratives such as in this/that way or like this/that. As it is mentioned, demonstrative adverbs will not be the main interest in this study.

2.4 Demonstrative Identifiers

A relatively unfamiliar category of demonstratives is demonstrative identifiers, also known as “identificational demonstratives.” These resemble demonstrative pronouns in sentences, as in the two examples from French and Poanpean in (5) and (6) (Diessel 5).

(5) French

C’ est Pascal.

this/it is Pascal

‘It/this is Pascal.’

(6) Ponapean

Iet noumw pinselen.

here.is your pencil

‘Here is your pencil.’

Even though they are often considered demonstrative pronouns, these demonstratives are used in copular and nonverbal clauses and many languages distinguish them from demonstrative pronouns by using special demonstrative forms.2

1As much as this is an intriguing topic to look at, it will not be discussed in too much detail here since the focus of

the study is not on the difference of demonstratives and articles. However, for further understanding of the argument, read Van Valin and LaPolla’s work.

2For better understanding of demonstrative identifiers, read Diessel’s work in Demonstratives: Forms Function, and Grammaticalization.

Chung 4

Although the last two demonstrative categories, demonstrative adverbs and identifiers, are important, only demonstrative pronouns and determiners, counterpart to this and that in English, will be scrutinized in this study, especially when in regard to five major languages in this paper (English, Chinese, Spanish, Korean, and Japanese) mentioned later.

3. Pragmatics

Demonstratives are significant element in communication, serving many pragmatic functions. The two main uses of demonstratives are exophoric use and endophoric use, with endophoric use further divided into anaphoric, discourse deictic, and recognitional uses (Diessel 93) as seen in figure 1 below.

According to Lenz, exophoric demonstratives, which he simply terms as “real deixis”, and endophoric demonstratives, which he terms as “anaphora”3

, are “distinguished by the domain … in which the referent is accessible”. Reference to a physically present object is exophoric, whereas, reference to a textually present object, mentioned in the past, is endophoric (Lenz 225). In other words, exophoric demonstratives directly refers to an object or entity in the outside world that is newly introduced in speech, whereas, endophoric demonstratives bring back an object or entity that has been mentioned before.

This is a crucial distinction that relates to one of the principal questions stated in the introduction (i.e. the usage of demonstratives for entities outside of physical perception). In the following sections, differences in these two uses of demonstratives will be described in detail.

3Lenz neither uses the term endophoric nor subdivides it into the three uses introduced by Diessel. Instead, he

distinguishes the pragmatic uses into real deixis and anaphora. However, his discriminatory criteria for real deixis and anaphora is very close to Diessel’s criteria for exophoric and endophoric, and his description of anaphora seems to cover the most characters of endophoric.

Figure1. The pragmatic uses of demonstratives

Chung 5

3.1 The Exophoric Use

The exophoric use of demonstratives is probably the most widely and easily comprehended use of demonstratives. There are three features that make exophoric demonstratives different from endophoric demonstratives. First, someone, usually the speaker, becomes the deictic center. Second, except in the rare case of distance-neutral demonstratives, there is a distance scale that helps the speaker indicate a deictic contrast. Third, a pointing gesture often accompanies (Diessel 94). Fillmore introduces two distinct uses of exophoric demonstratives4

3.1.1 Gestural and Symbolic Uses

, which are gestural and symbolic uses.

The sentences in (7) and (8) contrast the gestural use and the symbolic use of this.

(7) This nail just broke.

(8)This country is so beautiful.

For gestural use of demonstratives (7), in order to know what the demonstratives are referring to, the addressee must know where the speaker is pointing. Whereas, for symbolic use in (8), the speaker cannot point at the place he is referring to. This in (8) “draws on knowledge about the larger situational context, which involves more than what is immediately visible in the surround situation” (Diessel 94).

To clarify this dependence with gestural demonstratives, imagine having a conversation with a blind person. It is apparent that with a blind person, gestural demonstratives lose their meaning. This explains why exophoric demonstratives are often described as “pointers” which locate an object or entity in the physical world.

In many situations, beyond physical gestures, complicated inferences are also required, e.g., in the situation provided by Clark et.al. in (9),

(9) (Margaret, gesturing in the direction of two newspapers Duncan is holding, asks Duncan) “Could I look at that newspaper?”

The usage of that in this situation seems odd since Duncan is holding two newspapers and that is singular. Even though Margaret is gesturing, from a distance the gestural direction can seem very ambiguous. However, if one of the newspapers has a significantly bigger headline, say “War Over,” then the usage of that might be natural and Duncan will know which newspaper Margaret is referring to (Clark 246).

4Fillmore divides demonstratives into three uses: gestural, symbolic, and anaphoric uses. Among these, gestural

and symbolic uses have the features of exophoric uses according to Diessel (94); therefore they are included under section 3.1 The Exophoric Use.

Chung 6

The symbolic use shows that exophoric demonstratives can also be used for referents without concrete or visible physical bodies in the surrounding situation. Also, if entities that do not have a physical existence are taken into account, such as this in (9),

(9) This is a depressing feeling,

then it is quite clear that the “pointers” description of exophoric demonstratives is too simplistic.

3.2 The Endophoric Use

Unlike the exophoric demonstratives, which point to the entities in the outside world, endophoric demonstratives serve a language-internal function that point to the entities in the discourse world. In regard to the discourse world and the endophoric use, Lenz states,

The discourse world is not independent of a ‘physical world’, but it is neither identical with the latter nor a subset of it: while talking to somebody, I can see people, trees, cars etc. without them playing any role in the discourse. By means of a deictic reference, however, they can be included in the discourse world at any time. Thus, on the one hand, a ‘discourse world’ is a selection of the things from the ‘physical world’ that are relevant to discourse. On the other hand, things which are not (any longer) physically perceptible, but which the speaker and/or hearer remember, can be (re-)introduced into the discourse world. (Lenz 225)

The latter description of the discourse world leads to an important concept termed Common Ground in this paper which will be discussed in a later part of this study. It is also closely related to the following sub-uses of endophoric use.

3.2.1 The Anaphoric Use

The main function of anaphoric demonstratives is to track participants of the preceding discourse, which means that the interlocutors must have common access to a previously mentioned object or entity that is now a referent. In an example given by Kuno in (11), the expression sono hito ‘that person’ in the second sentence is co-referential with a noun phrase in the first sentence.

(11) Japanese (Kuno 284)

Kinoo Yamada to yuu hito ni aimasita. Yesterday Yamada as named person met ‘Yesterday, I met a man by the name of Yamada.’

Sono hito, miti ni Mayotte komatte.ita node, tasukete agemasita. That person way in lose was.in.trouble because helping gave (the favor of) ‘Since he lost his way and was having difficulties, I helped him.’

Chung 7

Diessel also points out that even though some believe that the use of anaphoric demonstratives after the first appearance of a new discourse participant is more common in languages without a definite article, i.e. the in English, there is no difference based on the existence of definite article in a language. And “once a new discourse participant has been established as topic, it is usually tracked by third person pronouns, zero anaphors, definite articles, or pronominal affixes on the verbs; but when a referent is mentioned for the second time, demonstratives are often the most common tracking device” (98).

3.2.2 The Discourse Deictic Use

The discourse deictic use is the same as the anaphoric use in most parts, except that discourse deictic demonstratives are not co-referential with a prior NP. Rather than nouns or noun phrases, discourse deictic demonstratives refer to propositions and focus on the aspects of meaning expressed by “a clause, a sentence, a paragraph, or an entire story” (Diessel 101) as the example given by Webber in the following dialogue in (12):

(12) A: Hey, management has reconsidered its position. They’ve promoted Fred to second vice president.

B: a. That’s false. (reference to proposition)

b. That’s a lie. (reference to illocution) (Webber 111)

In this example, that in both (12a) and (12b) refers to the preceding utterance, however, that in (12a) refers to the propositional content of the preceding utterance, while that in (12b) refers to the speaker’s linguistic act. Unlike anaphoric demonstratives, which track prior discourse participants, discourse deictic demonstratives connect two propositions –the proposition where the demonstrative is embedded, for instance (12B), and the proposition which the demonstrative is referring to, (12A).

Another difference between the anaphoric use and the discourse deictic use is that, while anaphoric demonstratives can only refer to a previously introduced entity, discourse deictic demonstratives can refer to both elements of the preceding discourse or the following discourse. The sentences in (13) and (14) illustrate these possibilities:

(13) A: I’ve heard you will move to Hawaii?

B: Who told you that (*this)?

(14) A: Listen to this (*that): John will move to Hawaii. (Diessel 102)

Chung 8

Although in (13) and (14), the demonstratives used are different5

3.2.3 The recognitional use

, both demonstratives function as discourse deictic demonstratives referring to the proposition of John moving to Hawaii, regardless of their occurrence before or after the relevant discourse.

The two endophoric uses above, anaphoric and discourse deictic, involve discourse old information, i.e. information that is shared and activated at some point in discourse. Recognitional demonstratives refer to information that is unactivated (discourse new) but pragmatically presupposed (hearer old). Consider the example below in (14):

(15) … it was filmed in California, those dusty kind of hills that they have out here in Stockton and all, … so… (Himmelmann 230)

In this example, dusty hills is not mentioned anywhere before, but instead of leaving the slot empty, a demonstrative those is used. This indicates that the following noun or noun phrase, in this example, dusty hills, expresses information familiar to the hearer due to specific shared knowledge – which is called Common Ground in this paper.

It is important to note that the information referred by recognitional demonstratives is not only discourse new and hearer old, but also private. Private information means that the information is shared by the speaker and the hearer through a common experience in the past. Unlike private information, general cultural information, even if it is discourse new and hearer old, is usually marked by a definite article in English. Recognitional demonstratives are often called “emotional deixis”, because of their function as an indicator of emotional closeness, sympathy, and shared beliefs (Lakoff 351)

3.3 Summary

Table1 below summarizes some of the essential points of the four uses of Demonstratives:

Table 1. Pragmatics of Demonstratives Exophoric Endophoric

Anaphoric Discourse Deictic Recognitional --- After referent Before/after referent Immediately before referent –

only used as determiner (adnominal)

In the outside world Language-internal Language-internal --- --- Discourse old Discourse old Discourse new, hearer old,

private

5 See further discussion in the later section5.1.

Chung 9

4. Semantics

Among the three semantic categories of deixis – person, place, and time (seeTable 2), demonstratives are place deictics. In other words, demonstratives express the spatial relation of the referred entity to the deictic center, which is usually the speaker. In this section, various interpretation of place demonstratives will be described.

Table 2. Examples of person, place, and time deixis in English

Person I, you, he

Place this, that, these, those

Time yesterday, today, now

4.1 Distance

When they are asked to explain the difference between the contrasting demonstratives such as this/these and that/those in English, distance is the first thing most people think of. In the majority of languages, like English and Chinese, two-term systems are used. In such languages, there are two contrasting demonstratives, and they are understood and used based on a distinction between proximal and distal (Huang 152), see Table 3.

Table 3. English and Chinese Demonstratives

Proximal Distal

English this/these that/those

Chinese zhe Na

Other languages, such as Spanish, Korean, and Japanese, use three-term systems, and these systems are either distance-oriented or person-oriented.

Distance-oriented languages understand their three demonstratives with a three-way contrast: proximal (close from the speaker), medial (far from the speaker), and distal (extremely far from the speaker and/or the hearer). In this system, the middle term is relative to the deictic center, typically the speaker. Spanish is one of the languages that use the distance-oriented three-term system6

6Some scholars such as Huang do not agree that Spanish belongs to the distance-oriented category; rather, they

believe that it is a language whose three-term system is not easily defined as either distance-oriented or person-oriented (154). However, other scholars believe Spanish to have a distance-oriented system (e.g. Diessel 39).

, see Table 4.

Chung 10

Table 4. Spanish Demonstratives

Proximal Medial Distal

Spanish este/estes esta/estas

ese/esos esa/esas

aquel/aquellos aquella/aquellas

On the other hand, person-oriented languages have the middle term relative to the addressee. In these languages, three demonstratives are understood as proximal to the speaker, proximal to the addressee, and distal from both the speaker and the addressee. Korean and Japanese both use such system, see Table 5.

Table 5. Korean and Japanese Demonstratives

Proximal (S) Proximal (A) Distal (S+A)

Korean i- ku- ce-

Japanese ko- so- a-

There are still other languages that have four or more than four-term system, and they vary with their ways of contrasting the terms; some incorporate audience (i.e. inactive participants in the conversation), while some regard distance proximal to both the speaker and the addressee (Huang 157). However, since the five languages studied in this paper have either two or three-term systems, systems with four or more terms will not be discussed in depth.

4.2 Other Deictic Parameters: Visibility, Elevation, and Motion

There are other deictic parameters discussed by scholars (e.g. Huang 156; Diessel 41). The five languages in this study are not thought to have demonstratives distinguished by these other parameters (visibility, elevation, and motion); however, the latter will reappear in the later sections. Thus, a brief mention of them will be made.

4.2.1 Visibility

Other than the well known distance parameter of deixis, visibility is another interesting deictic element, which is commonly seen in Native American languages (Diessel 41). For instance, as shown in Table 6, Ute has demonstratives for proximal, distal, and invisible (i.e. out of sight) referents.

Chung 11

Table 6. Inanimate Nominal Demonstratives of Ute7

Proximal Distal Invisible/out of sight

Ute ica Maru Uru

The invisible demonstratives can be specified by being further categorized, for instance, near (partly visible), known place, and unknown place in Quileute, or invisible-remote, invisible-occlusion, and invisible-periphery in Imai.

4.2.2 Elevation

Elevation is another dimension of demonstratives where distinctions are made based on the relative height of the referent. Languages with this parameter use geographical features such as uphill and downhill or upriver and downriver, or make the deictic center set a horizontal line as the zero point from where being upward or downward is decided (Huang 159).

4.2.3 Motion

Also called “stance” by Huang, demonstratives with a motion parameter indicate the movement or direction of the referent relative to the deictic center. For instance, whether the referent is standing, sitting, coming, or going (158).

5. Survey on the Usage of Demonstratives

With the above characteristics of demonstratives as backward the actual usage of demonstratives in five languages – English, Chinese, Spanish, Korean, and Japanese – was studied through an online survey. This section will present the objective and hypotheses of the survey regarding the usage of demonstratives (5.1), recapitulate characteristics of demonstratives in the five languages and provide the demographics of the participants (5.2), present the survey method (5.3) and its results (5.4).

5.1 Objectives and Hypotheses

The objective and the interest of the survey were to find out whether there are factors other than the physical distance between the referent and the deictic center (e.g. proximal and distal) that govern the choice of demonstratives in the speaker. Five languages, see section 5.2, were used to see if the potential factors would have the same degree of influence in different yet similar languages. Due to

7Demonstratives of Ute are also distinguished by animacy, which is one of the qualitative features of

demonstratives presented by Diessel (47). However, qualitative features are not necessarily relevant to the interest of this study, thus, will not be included in the sections.

Chung 12

time constraints and the fact that it was the first survey designed for the study, many variables were tried at once. The factors addressed in the study– or mini-hypotheses – were:

(a) Common Ground: As pointed out in the endophoric use of demonstrative, especially the recognitional use in section 3.2.3, shared knowledge or experience between interlocutors have an important pragmatic function in the use of demonstratives. And the presence or absence of Common Ground, i.e. whether the addressee has any prior knowledge about the referent, might influence the choice of usage between this and that.

(b) Source of Idea: As examples (13) and (14) show in the discourse deictic use in section 3.2.2, where the proposition originated might make a difference in the usage of demonstratives. Consider the English demonstratives used in (13’) and (14’) below:

(13’) A: I’ve heard you will move to Hawaii?

B: Who told you that (*this)?

(14’) A: Listen to this (*that): John will move to Hawaii.

In (13’), the speaker is referring to an idea brought up by the second person “you”, whereas, in (14’), the speaker (the first person “I”) is presenting the idea as if it was unheard of before and is originated from him. Although the referent (the proposition) is identical, the choice of demonstratives is very clear and different. This is possibly related to the person-aspect of demonstratives (see person-oriented system in section 4.1) or the time frame, i.e. when is the reference made.

(c) Perception: Referring back to section 4.2.1, physical perception such as vision is often incorporated into the talk of demonstratives; it is easy for people to use visible and tangible objects as referents. However, sometimes only other perception such as olfaction and audition can be used to perceive the referent, e.g. referring to the noise one hears. Referents that are invisible yet still perceived through other means might have different influence on the choice of demonstratives from concrete objects out of sight.

(d) Distance from the Head: The height parameter in section 4.2.2 alludes to the fact that in languages where height is taken into consideration, a horizontal line marked as the zero point is set by the deictic center, say the speaker. The question is: does the speaker as a whole mark this zero point, or is there a more specific part of the speaker that functions as the deictic center? If the latter proposal is true, the head of the speaker, where his eyes are, might be a potential part where the deictic center is positioned. Therefore, objects right above or on the head and objects below or on the feet might show a difference in terms of its distance from the head. In addition, whether the referent has a bodily contact to the speaker might also have an effect.

The general hypothesis of the study was that the factors, or mini-hypothesis, (a)-(d) above, will impact the usage of demonstratives; a possible pattern is sought in the survey results.

Chung 13

5.2 Languages of the Study

The five languages were initially chosen based on the investigator’s familiarity with them; two of the languages (English and Chinese) with two-term system demonstratives, and three of the languages (Spanish, Korean, and Japanese) with three-term system demonstratives. According to Hofmann, “superficially, the systems of English, Spanish, and Japanese are rather different … Nevertheless, when we look closely at the semantic elements underlying these systems we find amazing similarities…” (70). Therefore, English, Spanish, and Japanese– plus Chinese and Korean, whose demonstratives are also very similar to those of English and Japanese, respectively– are appropriate selection of languages to juxtapose for comparisons.

5.2.1 English: This and That

English demonstratives, though originally a three-term system with the third term yonder, have developed into a two-term system. The contrasting terms this/these and that/those are defined as proximal and distal (Table 3 in section 4.1), and are in the pronoun and determiner categories of demonstratives (see section 2).

There were 9 English speakers for the English portion of the survey. All but one participant, who was born in Uzbekistan with Russian as her native language, were born in the United States, and all of their currently preferred languages were English.

5.2.2 Chinese: Zhe and Na

Chinese, like English, has recently changed from a three-term system to a two-term system (Hofmann 65). Also like English demonstratives, Chinese demonstratives are distinguished as proximal and distal (Table 3 in sectin 4.1).

There were 15 Chinese speakers for the Chinese survey. All of them were born in either Hong Kong, Taiwan, or mainland China, all have some dialect of Chinese (Mandarin or Cantonese) as their native language, and all but one, who preferred English, indicated that their currently preferred languages are Chinese, with 3 of them preferring English and Chinese equally.

5.2.3 Spanish: Este, Ese, and Aquel

Spanish has three demonstrative terms with a distance-oriented system: proximal, medial, and distal (Table 4 in section 4.1). Some believe that Spanish demonstratives have neither the distance-oriented system nor the person-oriented system, but it seems agreeable to say that Spanish demonstratives do have a different system from Japanese demonstratives. Also, unlike the middle term in Japanese, which is almost always translated as that in English, the middle term in Spanish is sometimes translated as this and sometimes translated as that (Lenz 15), hence distance-oriented system more fitting.

Chung 14

There were 6 Spanish speakers for the Spanish survey. All of them were born in Spain and Spanish is all of their native and currently preferred language.

5.2.4 Korean: I-, Ku-, and Ce-

Korean demonstratives are distance marking particles rather than independent words like in English, Chinese, and Spanish above. The particles i-, ku-, and ce- (see Table 5 in section 4.1) are followed by a noun or a noun phrase and function as determiners. Unlike in the other four languages, Korean demonstratives cannot be used as demonstrative pronouns unless they are followed by a defective noun, an otherwise meaningless word indicating the type of referent when combined with the demonstrative particle (Sohn 294). For instance, (15), an example from Sohn, shows the proximal particle i- prefixed to kes, which indicates ‘thing or fact.’ The combination create a pronominal demonstrative.

(15) i kes this thing/fact ‘this (one/thing/fact)’

There were 11 Korean speakers for the survey. All of them except two participants, one born in China and another born in Germany, were born in Korea, all of their native languages were Korean, and all of their currently preferred languages were Korean, with 3 of them preferring English and Korean equally.

5.2.5 Japanese: Ko-, So-, A-

Similar to Korean, Japanese demonstratives are distance-marking particles that cannot stand on their own. Japanese, unlike Korean, has two category markers that allow Japanese demonstratives to function as pronouns and determiners.

Table 5. Japanese Demonstratives: Pronouns and Determiners

Demonstrative Pronouns

Demonstrative Determiners

Proximal (S) ko-re ko-no Proximal (A) so-re so-no Distal (S+A) a-re a-no

There were 8 Japanese speakers for the survey. All of them were born in Japan, all but one, whose native language is English, listed Japanese as their native language, with one of them listing both Japanese and English as the native language, and for the currently preferred language, 5 participants listed Japanese while 3 participants listed English.

Chung 15

5.3 Methods of Study

5.3.1 Material

A set of questionnaires were created in English and translated into the four other survey languages (see Appendix II for questionnaires in English). The set consisted of 15 questions, in which two of the questions were targeting Common Ground (questions 8 and 9 in Appendix II), four targeting Source of Idea (questions 11 to14), one targeting Perception, and the rest targeting Distance and Bodily Location (questions 1 to 7 and 10). Some of the questions regarding Distance were present to confirm the characteristics of demonstratives in the sections above (for summarized features of each question see Appendix I). Simple pilot testing was done for Chinese, Spanish, and Korean translation.

The questions were designed in such a way that differing variables, the targeting factors, would not overlap and confound the result; however, Source of Idea inevitably overlapped with features of Common Ground. This is because in a first person originated idea –often an innovative thought presented or introduced into the discourse for the first time by the speaker, it is natural that there is no shared knowledge or experience of the referred idea yet (see question 3 in Appendix II); while in a second, third, or unknown-person-from-the-past originated idea – where idea has to be referred by the speaker sometime after its mention, there is already a Common Ground established.

The questionnaires were multiple-choice questions asking the participants to fill in the blank with the appropriate choice(s), samples from English survey and Spanish survey are provided in Figure 2.

In the English and Chinese surveys, participants were allowed to choose only one choice – this, that, or either this or that; whereas, in Spanish, Korean, and Japanese surveys, participants were allowed to choose more than one possible answer to avoid the complication of giving them answer choices such as either esta or esa and either esa or aquella.

5.3.2 Procedure

Participants were recruited and their email addresses were collected via classroom announcements and acquaintance requests. An online survey request was sent to the recruited people from an online survey program (SurveyMonkey.com), and a total of 49 participants responded to the survey request.

Figure 2. Samples of English and Spanish Surveys

Chung 16

Figure 3. Common Ground Results in English and Chinese

Prior to answering the actual questionnaires, participants were asked to electronically sign a consent form and answer three short-answer questions regarding their language background for demographic purposes. They were asked to provide their place of birth, native language, and currently preferred language. For the questionnaires themselves, the participants were required to read the sentences carefully, imagine that they are the character speaking, and select either (i) the best choice, for English and Chinese surveys, or (ii) all possible choices, for Spanish, Korean, and Japanese surveys. After each question items, they were given an optional short-answer question regarding their reasons for the choice they made8

5.4 Results

.

5.4.1 Common Ground

Question 1 and 2 were designed to see the contrast between the presence and absence of Common Ground (Appendix II).

In the English and Chinese surveys, when Common Ground was present, distal demonstrative was selected by most English (88.9%) and Chinese (44.4%) speakers. When Common Ground was absent, the selection was almost equal for proximal and distal demonstratives in English, while distal demonstrative was selected by all Chinese speakers– see Figure 3 below.

In the Spanish, Korean, and Japanese surveys, when Common Ground was present, middle9

8 For the Korean survey, the indication that this short-answer question is optional was mistakenly left out. As a

result, almost all Korean participants gave all the reason for their choices.

9The second demonstrative term is labeled as “middle” here instead of medial, because in Korean and Japanese, the middle term indicates that the referent is proximal to the addressee rather than having a medial distance it the speaker, like in Spanish.

and distal demonstratives were almost equally selected by Spanish speakers, middle demonstratives were selected by all Korean speakers (100%), and distal demonstratives were selected by all Japanese

Chung 17

Figure 4. Common Ground Results in Spanish, Korean, and Japanese

speakers And when Common Ground was absent, there was no change in the selection of choices in Spanish and Korean speakers, while middle demonstratives were selected most often by Japanese speakers (83.3%)– see Figure 4 below.10

Comments from participants indicate that many of them chose the non-proximal demonstrative because the referent was someone met in the past. This is probably the reason Spanish and Korean results did not differ based on the existence of Common Ground. Time frame seems to play a crucial role here and possibly confounded the result.

5.4.2 Source of Idea

Questions 3 to 6 were designed to reveal the difference made by first person, second person, and third person ideas, and question 6 was for an unknown person in the past (Appendix II).

In the English and Chinese surveys, when the referred idea was from the first person, the speaker, proximal demonstrative was selected by most English (100%) and Chinese (93.3%) speakers; when the referred idea originated with the second person, the addressee, proximal demonstrative was selected by all English speakers, while proximal and distal demonstratives were equally selected by Chinese speakers. When the referred idea was from the third person, someone not directly in the conversation, the distal demonstrative was selected by most English speakers (77.8%), while proximal demonstrative was selected by most Chinese speakers (80%). Finally, when the referred idea was from an unknown person in the past, distal demonstrative was selected by most English (88.8%) and Chinese speakers (80%) – see Figure 5 on the following page.

In the Spanish, Korean, and Japanese surveys, when the referred idea was from the first person, the speaker, the proximal demonstrative was selected by most Spanish (100%), Korean (100%), and 10The total number of percentage might seem odd in the Spanish, Korean, and Japanese data. Unlike English and Chinese data, where the total percentage makes up a 100%, Spanish, Korean, and Japanese data sometime give total percentage more than a 100%. This is because in Spanish, Korean, and Japanese surveys, participants were allowed to choose more than one choice.

Chung 18

Japanese (83.3%) speakers. When the referred idea was from the second person, the addressee, the middle demonstrative was selected by most Spanish speakers (83.3%), all Korean speakers and Japanese speakers. Half of the Spanish speakers (50%) also either selected the proximal or both the proximal and distal demonstratives. When the referred idea was from the third person, someone not directly in the conversation, the proximal was selected the most by the Spanish speakers (66.7%), the middle demonstrative was selected the most by Korean (50%) and Japanese (66.7%) speakers. Finally, when the referred idea was from an unknown person in the past, the distal demonstrative was selected by most Spanish (83.3%) and Japanese speakers (85.7%), with the middle demonstrative also being selected by quite a few Spanish speakers (33.3%), and the middle demonstrative selected by most Korean speakers (90%), with the proximal demonstrative also being selected by some of them (30%) – see Figure 6 on the following page.

Comments from participants for questions 3 to 6 indicated that where the referred source originated did matter in their choice. They also pointed out that temporal distance, e.g. how far away in the past was the idea first mentioned, and Common Ground, the location of the original source, i.e. whether the person who thought of the idea was present in the conversation, did influence their choices (see Appendix III).

Chung 19

Figure 5. Source of Idea: English and Chinese

Figure 6. Source of Idea: Spanish, Korean, and Japanese

Chung 20

Figure 7. Perception: English and Chinese

Figure 8. Perception: Spanish, Korean, and Japanese

5.4.3 Perception Questions 7 was designed to see how a referent perceived by olfactory sense alone affects

demonstrative choices (Appendix II).

In the English and Chinese surveys, when the referent, a smell in this case, was invisible and was perceived only by olfaction, the distal demonstrative was selected by most English speakers (77.8%) while the proximal demonstrative was selected by most Chinese speakers (93.3%) – see Figure 7 below.

In the corresponding Spanish, Korean, and Japanese surveys, the middle demonstrative was selected by most Spanish speakers (83.3%), with the proximal demonstrative also being selected by some of them (33.3%), while the proximal demonstrative was selected by most Korean (90%) and Japanese speakers (85.73%), with the distal demonstrative also being selected by many Korean speakers (50%) – see Figure 8 below.

Comments from participants show that time frame (that the speaker is experiencing the referred object at that time), Common Ground (that the speaker and the addressee are having the experience together), and the personal relevance (whether the referred object has any relation to the speaker (e.g. ownership)) became the reasons behind their choice of demonstrative.

Chung 21

Figure 9. Distance: English and Chinese Figure 10. Distance: Spanish, Korean, and Japanese

5.4.4 Distance

For questions regarding distance, results for four of the questions will be presented. Questions 8 and 9 were designed to see if the choice of demonstrative would differ when the referent is distal from both the speaker and the addressee but close to a participant (question 8) or no participant involved (question 9) – the latter makes the referent farther from the speaker and the addressee in these questions.

In the English and Chinese surveys, when the referent was distal but close to a participant present, the distal demonstrative was selected by all English and Chinese speakers (both 100%). When there was no participant involved and the referent is farther away, distal demonstrative was again selected by all English speakers (100%) and most Chinese speakers (86.7%) – see figure 9 below.

In the Spanish, Korean, and Japanese surveys, when the referent was distal but close to a participant present, the middle demonstrative was selected by most Spanish speakers (83.3%), the distal demonstrative was selected by all Korean speakers (100%), and all three demonstratives were selected by Japanese speakers (proximal 42.9%, middle 14.3%, and distal 57.1%) – see figure 10 below.

Chung 22

Figure 11. Bodily Location: English and Chinese

Figure 12. Bodily Location: Spanish, Korean, and Japanese

Questions 10 and 11 were designed to see if the referents’ distance from the head and bodily contact would influence the choice of demonstratives. The referent in question 10 was on the speaker’s knee, whereas the referent in question 11 was below the speaker’s feet.

In the English and Chinese surveys, when the referent was on the speaker’s knee, the proximal demonstrative was selected by most English speaks (66.7%) and all Chinese speakers (100%). When the referent was below the speaker’s feet without direct contact on the speaker’s body, the proximal and distal demonstratives were equally selected by English speakers and the proximal demonstrative was selected by most Chinese speakers (93.3%) – see Figure 11 below.

In the Spanish, Korean, and Japanese surveys, when the referent was on the speaker’s knee, the proximal demonstrative was selected by all Spanish speakers (100%) and most Korean (90%) and Japanese speakers (85.7%). When the referent was below the speaker’s feet without direct contact on the speaker’s body, a similar result was found: the proximal demonstrative was selected by all Spanish speakers (100%) and most Korean (80%) and Japanese speakers (100%) – see Figure 12 below.

Chung 23

Comments from participants indicate that other than the physical distance, personal relevance and relation with the object, e.g. how the speaker felt about the object, influenced their choice of demonstrative.

6. Discussion

The influence of Common Ground on demonstrative choice seems to be supported by English and Japanese results. When there was a Common Ground, most English and Japanese speakers selected the distal demonstrative, that and ano, respectively; but when there was no Common Ground, the number of English speakers selecting the distal demonstrative decreased and most Japanese speakers selected the middle demonstrative sono instead of the distal ano (section 5.4.1 Figure 3 and 4). In other words, an absence of Common Ground appears to decrease the mental distance of the referred object or idea from the speaker. When the Common Ground is present, the referent belongs to both the speaker and the addressee, but when it is not there, the referent belongs only to the speaker until he or she brings it out into the conversation. However, in cases of Chinese, Spanish, and Korean, it is speculated that the participants did not see the subtle difference in the questions, or there could have been other elements in play, such as time frame – distal and middle demonstratives were selected whether or not Common Ground existed because the referent was an entity present in the past.

In addition, the influence of Source of Idea in demonstrative choice seems to be supported by the results from English, Spanish, Korean, and Japanese. In all four languages, the speaker’s mental distance from the referred idea seems to increase as the Source of Idea becomes further (section 5.4.2. Figures 5 and 6). The contrast between 1st and 2nd person is the most clear and resembles the discourse deictic examples in (13) and (14). Comments from the participants clearly showed that source of idea does have an effect. Many described the introductor of the idea as being the “owner” of the idea; that is, in the 1st person case, the speaker would use a proximal demonstrative to, in a way, emphasize the fact that the referent belongs to him. The 3rd person case and the past unknown-source case seem to be more complicated. There could have been confounding factors such as time frame and presence of the 3rd person, who is the actual source of idea.

The data from Perception did not show any significant pattern (section 5.4.3). The comments gave mixed reasons for the choices participants made. The question might not have been worded in the best way to test the olfaction variable; comments indicated that some participants took the source of smell (the object that contains the smell) as the referent rather than the smell itself.

The data from Distance also did not show any significant pattern (section 5.4.4). However, it did confirm the possibility of Spanish being distance-oriented. As the referent becomes physically farther, the preferred Spanish demonstrative also changed from middle, which is medial in the distance-oriented system, to distal. When, compared to Spanish, the person-oriented demonstratives in Korean did not show any difference. The Bodily Location data showed that referent being on the speaker’s knee and being below his feet both made the participants choose proximal demonstratives. This is probably

Chung 24

because even when the referent was below the feet and not directly in contact with the body, it was still close to the speaker’s body in general. According to the participants, personal relevance, such as the impact the referent has on the speaker, and other factors seem to be present.

More interesting than the numerical data were the comments provided by the participants. Although the survey questions had many confounding factors and defects, ultimately, the objective was to find out the reason behind the speakers’ usage and choice of demonstrative. The participants’ understandings of themselves gave more valuable data than data from the inadequately manipulated questions. Among these valuable data, at least two other potential factors are found:

• Ownership: In Korean, the second demonstrative guh (in contrast to the third, juh) gives the impression that the addressee “possesses” the referred physical object, rather than a neutral object simply being near the hearer. Therefore, even when the referent is near the addressee, the third term might be preferred if the referent is not possessed or known by the addressee. (Supported by data such as comment (m) in Appendix III)

• Emotional distance: In Spanish, when referring to a person, the medial demonstrative (the second term) eso can denote a negative feeling, i.e. emotionally distance from the speaker (supported by comment (bb) and (nn)), while the distal demonstrative (the third term) aquel can denote the nonexistence of the person, e.g. the person is dead. 11

7. Conclusion and Future Studies

As was mentioned in the introduction, there is no specific guideline for the usage of demonstratives, and explanation based on physical distance alone is simplistic. Based on the above findings, more than one factor might be at work in a given situation that influences the choice of demonstratives. Some of the factor the participants identified in their comments that must be mentioned are:

a) tense (time frame or whether referent is an entity brought back from past experience),

b) ownership of the topic (source of idea or who started or suggested the idea),

c) shared knowledge (Common Ground or whether the referent is familiar to both interlocutors),

d) emotion and feeling toward the referent (personal belief or whether the speaker has any strong, especially negative, feeling toward the object) – which is very similar to Lakoff’s “emotional deixis” section 3.2.3, and

11 These two points are suggested from sources outside of the commentary data in Appendix III. The Korean one is suggested by one of the Korean participants after the survey, and the Spanish one is from Professor Ordóñez in the Stony Brook University Hispanic department.

Chung 25

e) personal meaning (how relevant the referent is to the speaker).

None of these factors is directly related to physical distance. However, distance still seems to remain a crucial part in the choice of demonstratives. As one of the participants noted, rather than the “distance closeness”, the “situational closeness” seems to play a major role in the choice of demonstratives (comment (nn) in Appendix III).

Perhaps, under a big category named “Distance”, there can be subcategories named “physical distance” and “mental distance,” which work together to determine which demonstrative would be the most appropriate and effective at delivering the right message in a given situation. Furthermore, under the two subcategories, there will be various features, such as the factors in (a)-(e) for “mental distance”.

For future studies on this topic, surveys may be accompanied by images to prevent the participants from having differing interpretation of the listed situation, especially when the referent is a physical object. It might also of interest to examine the three-term languages only, for instance, by comparing the appearance of medial and distal terms in Spanish or the second (proximal to the addressee) and third (distal from both interlocutors) terms in Korean and Japanese.

Chung 26

Appendix I: Variables in Survey Questions

Common Ground Source of Idea Perception Distance Bodily Location

1 Yes

2 No

3 No 1st person (S) 4 Yes 2nd person (A) 5 Yes 3rd person (P) 6 Yes Unknown (Past) 7

Olfactory

8

Distal (S+A); Proximal (P) Head (P) 9

Distal (S+A) ---

10

Proximal (S) Knee (S) 11

Proximal (S) Below body (S)

12

Proximal (S) Shoulder (S) 13

Proximal (A) Shoulder (A)

14

Proximal (S) Arm (S) 15

Proximal (S) Belly

S: Speaker

A: Addressee

P: Participant, i.e. an inactive audience in the conversation circle or simply present in the place where conversation is taking place

Chung 27

Appendix II: Survey Questionnaires in English

Questions for Common Ground:

1. Tom was talking to Jack about the girl they both met on Wednesday, “_____ girl was really pretty.”

2. Tom was talking to Jack about the girl he met on Wednesday, “_____ girl was really tall.”

Questions for Source of Idea:

3. During a meeting, Tom thought of an innovative idea and said, “hey, what about _____?”

4. Tom and Cindy were sitting next to each other planning the coming vacation, Cindy said they should go to a butterfly museum. And with a dislike, Tom said, “no, I don’t really like _____.”

5. Tom and Jack were in a meeting. As soon as Jack finished presenting his idea, Tom said to the boss referring to the idea Jack just presented, “yeah, what about _____?”

6. During a meeting, Tom thought of the idea they mentioned in the last meeting and said, “hey, what about _____?”

Questions for Perception:

7. When Tom and Deb walked into their dorm kitchen, they smelled someone’s brownie in the oven. Tom said to Deb, “_____ smells delicious.”

Questions for Distance and Bodily Location:

8. Tom, Deb, and Cindy were sitting at a round table, and while Cindy was listening to Tom and Deb’s conversation, a humungous fly flew by and sat on Cindy’s head. Knowing that Tom has noticed the fly too, Deb said to Tom, “do you see _____ fly on Cindy’s head?”

9. Tom and Deb were walking on the street, and a noise came from two blocks away. They turned their heads and saw two men fighting two blocks away. Deb said to Tom, “I think I’ve seen _____ man before.”

10. Tom and Jack were standing outside waiting for their ride, and Tom found a piece of gum stuck on his left knee. He said to Jack, “ew, when did _____ get on my knee?”

11. A world record balancer was walking on a rope from one side of the cliff to the other, and a blue bird came flying right below his feet. He said to himself, “gosh, I hope _____ bird flies away soon.”

12. Tom and Deb were walking side by side on a street, and a toy airplane began flying next to Tom’s shoulder. Tom said to Deb, “hm, I wonder who’s controlling _____ plane.”

Chung 28

13. Tom and Deb were walking side by side on a street, and a colorful butterfly came by and flew next to Deb’s shoulder. Tom said to Deb, “oh, _____ butterfly is so colorful.”

14. Tom found a ladybug on his arm and thought to himself, “look at _____ thing!”

15. Tom and Jack were standing outside enjoying the weather, and Tom found a random sticker on his t-shirt on his belly. He said to Jack, “I wonder where _____ came from.”

Chung 29

Appendix III. Comments from the Survey Comments from question 1-2:

a. “They both met her and it was in the past (not here and now)” b. “the more days after they met her, the more likely the use of "aquella"” c. “Because she is someone met in the past ku” d. “Because the girl is not in front of their eyes, but the incident is rather far from the time of

discussion” e. “Since he’s reflecting the past incident, ku is appropriate. If the situation were that he had the

girl’s picture in his hand, it would be pointing to a close by object so yi would be possible too.” f. “In my opinion, as the conversation goes on, we may replace this that with this (na)” –

participant has chosen a distal demonstrative na in Chinese g. “Because the speaker and the hearer are talking about a third person ku” h. “Because she is not someone they’re with”

Comments from question 3-6:

i. “Its not his idea” j. “Pointing at an idea he thought of himself” k. “not sure if the idea is from both Tom and Jack or only Jack” – selected zhe and na l. “Because it is Tom’s idea to bring up the question” – selected zhe m. “If he emphasizes that the idea is from himself i is appropriate, if he is trying to remind and use

someone else’s idea it can also be expressed with ku” n. “Because a new talk or topic is brought up” o. “Ku is appropriate because it is pointing at the subject suggested by Cindy” p. “More immediate” q. “They just talk about the idea before.” r. “Hinting what he is going to talk about” s. “If he has the document in his hand i is possible, calling the subject from the past so ku is

appropriate” t. “Ku is correct since the time passed” u. “Because they’re talking about something mentioned in the past” v. “Because they are talking about something talked about before and they are talking about the

same topic” w. “The discussed matter is something both of them know” x. “both of them commonly knowing, ambiguous call” y. “The content is not certain but something both should know” – selected ku z. “if he refers to the vacation plan, he can use "esto". If thinking only about the museum, he can

use "eso"” aa. “if he says that with Jack present, he would use "esto". If talking to the boss privately, the other

possibilitites also apply”

Chung 30

bb. “Wants to distance it because it’s something he hates”

Comments from question 7: cc. “He doesn't necessarily have ownership to the brownies but he wants to” dd. “If there is a direct relation with the smell then i-smell, if referring to the oven exactly then ce-

smell” ee. “It can be i since the addressee is close by, it can be ce since the person is not of relevance to

the speaker and the hearer” ff. “They are speaking while they are smelling.” (time frame) gg. “The smell they are feeling right now.” (Close) hh. “Because its commonly sensed, two people have the same experience” (common ground) ii. “To expressed the smell itself i is appropriate, if looking at the brownie and placing the focus on

the browie, ce is appropriate” Comments from question 9:

jj. “the mind/focus brings closer the subject than it is in reality, that's why he can say "ese" while the norm would be "aquel" (in my opinion)”

Comments from question 10:

kk. “Tom wants to put distance between him and the gum, because its gross” ll. “Because he doesn’t know where the gum came from ce”

Comments from question 11:

mm. “Since it is close because it is under the person’s feet it can be i, but since it is an object irrelevant to the speaker it can also be ce”

nn. “Talking to oneself, it’s a relation only between the balancer and the bird? He is not giving a direction to anyone, besides, it is situated in a close enough distance (situational closeness rather than the distance closeness)”

oo. “Because there is no hearer and is talking to himself, i is appropriate” Comments from question 13:

pp. “Even though the butterfly is close to the hearer, ku is usually used when it is related to the

hearer”

Chung 31

Reference

Clark, Herbert H., Robert Schreuder and Samuel Buttrick. "Common Ground and the Understanding of Demonstrative Reference." Journal of Verbal Learning and Verbal Behavior 22 (1983): 245-258.

Diessel, Holger. Demonstratives: Form, Function, and Grammaticalizatoin. Amsterdam: John Benjamins Publishing Company, 1999.

Fillmore, Charles J. Lectures on Deixis. Stanford: CSLI Publications, 1997.

Himmelmann, Nikolaus P. "Demonstratives in Narrative Discourse: A taxonomy of universal uses." In Fox (ed.) (1996): 205-254.

Hofmann, Thomas R. Realms of meaning: an introduction to semantics. New York: Longman Publishing, 1993.

Huang, Yan. Pragmatics. New York: Oxford University Press, 2007.

Kuno, Susumu. The Structure of the Japanese Language. Cambridge: MIT Press, 1973.

Lakoff, R. "Remarks on this and that." Chicago Linguistic Society 10 (1974): 345-356.

Lenz, Friedrich. Deictic Conceptualisation of Space, Time and Person. Amsterdam: John Benjamins Publishing Company, 2003.

Sohn, Ho-Min. Korean. London: Routledge, 1994.

Van Valin, R.D. Jr. and R.J. LaPolla. Syntax: Structure, meaning, and function. Cambridge: Cambridge University Press, 1997.

Webber, Bonnie Lynn. "Structure and Ostension in the Interpretation of Discourse Deixis." Language and Cognitive Processes 6

Acknowledgement

(1991): 107-135.

• Professor Richard Larson for being such a great advisor encouraging and supporting the research. • Professor Agnes He, Professor Francisco Ordóñez, Professor Eriko Sato, Jisung Sun, and Susana

Huidobro for helping with the translation, providing helpful literature, and finding participants. • Professor Gary Mar, Professor Robert Hoberman, and Professor Lori Repetti for being on the

committee and showing constant interest and support. • Stony Brook Undergraduate Research and Creative Activities (URECA) for funding the research.